4 ms·

I am writing my master thesis on free & open source mathematical tools. To my shame I have never heard of SageMath, it never came up on my radar. I had heard of

I am writing my master thesis on free & open source mathematical tools. To my shame I have never heard of SageMath, it never came up on my radar. I had heard of Numpy, scipy, matplotlib, maxima, Octave and others. I know that custom software is written in Fortran and C/C++- Why had I not heard of SageMath? Are there any other large open source mathematical projects that may have escaped my notice? Please enlighten me. The aim of my thesis is to discover why proprietary tools like Maple, Mathlab, and Mathematica have flourished within the mathematics world given that theorem proving and calculating seem to require access to the algorithms for verification processes. I aim to interview the leaders behind both proprietary and open tools to see what drives and motivates them. This video by William Stein http://www.sagemath.org/help-video.html http://www.sagemath.org/help-video.html on the SageMath website goes a long way towards answering some of my questions. I would like to look at this problem space from all angles, economic, ethical, scientific, you name it. I would welcome ideas on research strategies and any and all pointers. Speak up now, no matter how trivial you think your idea! Thank you tomrod for posting this, thank you.

hermanthegerman 16y ago
I used sage quite often when studying. I think many people today learn the one or the other programming language quite early, so it becomes natural to them, and an open source mathematical system like sage which extends their understanding of programming with mathematical/symbolic ideas is more powerful and way easier to adapt than those proprietory systems.
- igravious 16y ago
  That's an interesting way of looking at it. I always thought that the "perfect" mathematical tool should strive for parity with math standard math notation but the way you describe it makes perfect sense from the viewpoint of leveraging existing tools.
tomrod 16y ago
Here are some places I would immediately look--you most likely have come across some or all of these. I suspect some of the smaller proprietary products teams may be more accessible in interviewing. http://en.wikipedia.org/wiki/Comparison_of_computer_algebra_systems http://en.wikipedia.org/wiki/Comparison_of_computer_algebra_... http://en.wikipedia.org/wiki/List_of_numerical_analysis_software http://en.wikipedia.org/wiki/List_of_numerical_analysis_soft...
- igravious 16y ago
  I don't remember seeing these pages. Thanks. I have bookmarked: http://directory.fsf.org/category/math/ http://directory.fsf.org/category/math/ and http://en.wikibooks.org/wiki/FOSS_Education/Research_using_FOSS http://en.wikibooks.org/wiki/FOSS_Education/Research_using_F... There are a number of other sites dedicated to pre-university level education of mathematics but I was more thinking about my own experience with Maple, Mathematica, and Matlab and FOSS tools of a quality to substitute for these. Having listened to the SageMath intro video I would add Magna to that list I suppose but maybe it is narrower in scope. I need to know how the tools are categorized. I guess each mathematical field has its own set of tools. I guess Computer Algebra Systems (CAS) would be the general term for the symbolic math side of things and Automated Theorem Provers (ATP) for the proof theory side of things. Ignoring ATP for the moment I suppose you can subdivide CAS many ways. I wouldn't even know how to list the divisions in mathematics or even if each division had a corresponding type of mathematical tool.
tzs 16y ago
> The aim of my thesis is to discover why proprietary tools like Maple, Mathlab, and Mathematica have flourished within the mathematics world given that theorem proving and calculating seem to require access to the algorithms for verification processes. We don't have access to the algorithms that a human mathematician working unaided uses to do things like evaluate a tricky integral, simplify a complicated expression, and so on, yet we can still verify their proofs. I don't see why him using a computer program, proprietary or not, would make a difference. We verify the proof by checking the result, not the thought processes or computational processes that were used behind the scenes to come up with it.

hermanthegerman 16y ago

I used sage quite often when studying. I think many people today learn the one or the other programming language quite early, so it becomes natural to them, and an open source mathematical system like sage which extends their understanding of programming with mathematical/symbolic ideas is more powerful and way easier to adapt than those proprietory systems.

igravious 16y ago
That's an interesting way of looking at it. I always thought that the "perfect" mathematical tool should strive for parity with math standard math notation but the way you describe it makes perfect sense from the viewpoint of leveraging existing tools.

tomrod 16y ago

Here are some places I would immediately look--you most likely have come across some or all of these. I suspect some of the smaller proprietary products teams may be more accessible in interviewing. http://en.wikipedia.org/wiki/Comparison_of_computer_algebra_systems http://en.wikipedia.org/wiki/Comparison_of_computer_algebra_... http://en.wikipedia.org/wiki/List_of_numerical_analysis_software http://en.wikipedia.org/wiki/List_of_numerical_analysis_soft...

igravious 16y ago
I don't remember seeing these pages. Thanks. I have bookmarked: http://directory.fsf.org/category/math/ http://directory.fsf.org/category/math/ and http://en.wikibooks.org/wiki/FOSS_Education/Research_using_FOSS http://en.wikibooks.org/wiki/FOSS_Education/Research_using_F... There are a number of other sites dedicated to pre-university level education of mathematics but I was more thinking about my own experience with Maple, Mathematica, and Matlab and FOSS tools of a quality to substitute for these. Having listened to the SageMath intro video I would add Magna to that list I suppose but maybe it is narrower in scope. I need to know how the tools are categorized. I guess each mathematical field has its own set of tools. I guess Computer Algebra Systems (CAS) would be the general term for the symbolic math side of things and Automated Theorem Provers (ATP) for the proof theory side of things. Ignoring ATP for the moment I suppose you can subdivide CAS many ways. I wouldn't even know how to list the divisions in mathematics or even if each division had a corresponding type of mathematical tool.

tzs 16y ago

> The aim of my thesis is to discover why proprietary tools like Maple, Mathlab, and Mathematica have flourished within the mathematics world given that theorem proving and calculating seem to require access to the algorithms for verification processes. We don't have access to the algorithms that a human mathematician working unaided uses to do things like evaluate a tricky integral, simplify a complicated expression, and so on, yet we can still verify their proofs. I don't see why him using a computer program, proprietary or not, would make a difference. We verify the proof by checking the result, not the thought processes or computational processes that were used behind the scenes to come up with it.